HAL Id: hal-00682513
https://hal.archives-ouvertes.fr/hal-00682513
Submitted on 26 Mar 2012HAL is a multi-disciplinary open access archive for the deposit and dissemination of sci-entific research documents, whether they are pub-lished or not. The documents may come from teaching and research institutions in France or abroad, or from public or private research centers.
L’archive ouverte pluridisciplinaire HAL, est destinée au dépôt et à la diffusion de documents scientifiques de niveau recherche, publiés ou non, émanant des établissements d’enseignement et de recherche français ou étrangers, des laboratoires publics ou privés.
Experimental climate effect on seasonal variability of
polyphenol/phenoloxidase interplay along a narrow
fen-bog ecological gradient in Sphagnum fallax
Vincent Jassey, Geneviève Chiapusio, Daniel Gilbert, Alexandre Buttler,
Marie-Laure Toussaint, Philippe Binet
To cite this version:
Vincent Jassey, Geneviève Chiapusio, Daniel Gilbert, Alexandre Buttler, Marie-Laure Toussaint, et al.. Experimental climate effect on seasonal variability of polyphenol/phenoloxidase interplay along a narrow fen-bog ecological gradient in Sphagnum fallax. Global Change Biology, Wiley, 2011, 17 (9), pp.2945-2957. �10.1111/j.1365-2486.2011.02437.x�. �hal-00682513�
1
Experimental
climate
effect
on
seasonal
variability
of
1
polyphenol/phenoloxidase interplay along a narrow fen-bog ecological
2gradient in Sphagnum fallax
3Vincent EJ Jassey1, Geneviève Chiapusio1, Daniel Gilbert1, Alexandre Buttler2,3,4,
Marie-4
Laure. Toussaint1, Philippe Binet1
5
6
1
Laboratoire Chrono-Environnement, UMR CNRS 6249, UFR Sciences, techniques et
7
gestion de l‘industrie, Université de Franche-Comté, F-25211 Montbéliard cedex, France.
8
2
Laboratoire Chrono-Environnement, UMR CNRS 6249, UFR des Sciences et Techniques,
9
16 route de Gray, Université de Franche-Comté, F-25030 Besançon, France.
10
3
Ecole Polytechnique Fédérale de Lausanne EPFL, Ecological Systems Laboratory ECOS,
11
Station 2, 1015 Lausanne, Switzerland
12
4
Swiss Federal Research Institute WSL, Site Lausanne, Station 2, 1015 Lausanne,
13
Switzerland
14 15
Correspondance to Philippe Binet
16
Laboratoire de Chrono-Environnement, UMR CNRS 6249, UFR Sciences, techniques et
17
gestion de l‘industrie, Université de Franche-Comté, 4 place Tharradin, Montbéliard 25211
18
cedex, France
19
Tel: +33 3 81 99 46 89; fax: +33 3 81 99 46 61
20
E-mail address: philippe.binet@univ-fcomte.fr
21 22
Running title: phenol/phenoloxidase interplay in peatland
23 24
Key words: carbon cycle, climate warming, ecological gradient, open top chambers, peatland, 25
phenoloxidases, polyphenols.
26
2 Abstract
28
Extracellular phenoloxidase enzymes play an important role in the stability of soil carbon
29
storage by contributing to the cycling of complex recalcitrant phenolic compounds. Climate
30
warming could affect peatland functioning through an alteration of polyphenol/phenoloxidase
31
interplay, which could lead them to becoming weaker sinks of carbon. Here, we assessed the
32
seasonal variability of total phenolics and phenoloxidases subjected to 2-3°C increase in air
33
temperature using Open Top Chambers. The measurements were performed along a narrow
34
fen-bog ecological gradient over one growing season. Climate warming had a weak effect on
35
phenoloxidases, but reduced phenolics in both fen and bog areas. Multivariate analyses
36
revealed a split between the areas and also showed that climate warming exacerbated the
37
seasonal variability of polyphenols, culminating in a destabilization of the carbon cycle. A
38
negative relationship between polyphenols and phenoloxidases was recorded in controls and
39
climate treatments suggesting an inhibitory effect of phenolics on phenoloxidases. Any
40
significant decrease of phenolics through repeatedly elevated temperature would greatly
41
impact the ecosystem functioning and carbon cycle through an alteration of the interaction of
42
polyphenols with microbial communities and the production of extracellular enzymes. Our
43
climate treatments did not have the same impact along the fen-bog gradient and suggested that
44
not all the peatland habitats would respond similarly to climate forcing.
45 46
3 Introduction
48
Boreal peatlands currently represent a terrestrial sink of carbon with approximately one-third
49
of the world‘s organic carbon (390-455 Pg) (Gorham, 1991; Moore, 2002). The ability of
50
peatlands to store atmospheric carbon resides in the long-term accumulation of partially
51
decomposed organic matter. The accumulated peat is mainly dominated by remnants of
52
mosses of the Sphagnum genus, highly enriched in recalcitrant organochemical compounds
53
such as polyphenols (van Breemen, 1995; Verhoeven & Toth, 1995). Such compounds play a
54
role both through a polyphenolic network linked to cell walls which could directly preserve
55
Sphagnum-derived organic matter from degradation, and through the release of water soluble
56
phenolics which directly interact with the surrounding environment (van Breemen, 1995;
57
Verhoeven & Liefveld, 1997). Phenolics produced by Sphagnum have a potential inhibitory
58
effect on fungal and bacterial activity and/or on enzymes involved in organic matter
59
decomposition (Wetzel, 1992; Fenner et al., 2005; Opelt et al., 2007; Mellegard et al., 2009).
60
Among the diversity of enzymatic activities recorded in peat soils, only phenoloxidases –
61
mainly produced by fungi – are involved in the polymerization, depolymerisation and
62
transformation of both complex and simple phenolic compounds (Pind et al., 1994; Thormann
63
et al., 2002; Fenner et al., 2005; Baldrian, 2006; Sinsabaugh, 2010). However, acidic
64
conditions, waterlogging and low soil temperatures that occur in peat soils were recognized to
65
limit phenoloxidase activity (Pind et al., 1994, Williams et al., 2000; Freeman et al., 2001a, b;
66
Toberman et al., 2008, 2010). Thus, carbon sequestration in peatlands is thought to partly
67
result from a suppression of phenoloxidase activity (Freeman et al., 2001a, 2004).
68
The expected increase of air temperatures in boreal regions is predicted to lead to a
69
destabilization of peatland carbon stores (Smith et al., 2004; Strack, 2008). Owing to the
70
temperature regimes that currently constrain biological activities, climate warming may
71
significantly impact the stability of the carbon cycle of peatlands by the breakdown of its
4
recalcitrant organic matter and thus act on ―the enzymatic latch‖ (Freeman et al., 2001a,
73
2004). However, recent research on the effect of climate change on phenoloxidases highlight
74
equivocal results in peatlands (Laiho, 2006; Fenner et al., 2007; Toberman et al., 2008, 2010).
75
In regions without permafrost the most fundamental distinction among peatland types
76
is between bog and fen (Bridgham et al., 1998, 2001; Rydin & Jeglum, 2006). Bogs and fens
77
have been found to have different plant communities, hydrology, nutrient availability, and soil
78
chemistry (Bridgahm et al., 1998, 2001; Wheeler & Proctor, 2000; Rydin & Jeglum, 2006).
79
Owing to these differences in biotic and abiotic settings, bogs and fens are likely to differ in
80
their response to climate change, (Weltzin et al., 2000, 2001, 2003). Recently, Jassey et al.
81
(2011a) demonstrated that microorganisms (e.g. testate amoebae) and their interplay with
82
polyphenols varied along a short fen-bog gradient. Accordingly, an understanding of how
83
climate change modifies carbon cycling in peatlands by modifying the
84
polyphenol/phenoloxidase interplay in different ecological setting is essential to assess the
85
capacity of peatlands to continue to store carbon.
86
The aim of this study was to investigate the impact of experimental climate warming
87
on seasonal variation of polyphenols, phenoloxidases and their interplay in different
88
ecological settings. These factors were studied at two depths along the living Sphagnum shoot
89
on a short ecological gradient from a transitional Sphagnum-dominated poor fen to a
90
Sphagnum bog with more pronounced micro-topography. Temperatures were manipulated
91
using open-top chambers placed on half of the sampling plots, and compared with control
92
plots. We hypothesized that (1) seasonal variations of polyphenols, phenoloxidases and their
93
interplay would be different between the structurally more complex Sphagnum ―bog‖ habitat
94
and the more uniform poor fen, and (2) the warming effect would alter the seasonal variations
95
of these factors along the fen-bog gradient.
5 Materials and methods
97
Field site and vegetation
98
The study site is an undisturbed Sphagnum-dominated mire situated in the Jura Mountains
99
(The Forbonnet peatland, France, 46°49‘35‘‘N, 6°10‘20‘‘E) at an altitude of 840 m a.s.l. Cold
100
winters (on average -1.4°C) and mild summers (on average 14.6°C) characterize the site. The
101
annual mean temperature measured at the site over a one-year period from 5th November 2008
102
to 30th November 2009 was 6.5°C and the annual precipitation 1200 mm.
103
Samples of Sphagnum fallax were collected within homogeneous areas of S. fallax
104
carpet across two adjacent areas selected in relation to their wetness, soil micro-topography,
105
vegetation and assessment of sources and decay of organic matter according to Delarue et al.,
106
(2011). The first sampling area (called ―fen‖) was a transitional Sphagnum-dominated poor
107
fen with a relatively flat and homogeneous topography, characterized by a moss cover
108
dominated by S. fallax and by the lack of S. magellanicum. Vascular plants such as
109
Eriophorum vaginatum, Vaccinum oxycoccus and Andromeda polifolia were recorded in very
110
low abundance. Scheuchzeria palustris and Carex limosa occurred outside of the studied
111
plots. The second sampling area (called ―bog‖) was a Sphagnum bog directly adjacent to the
112
fen area. Patterns of hummocks with S. magellanicum, V. oxycoccos, E. vaginatum and
113
Calluna vulgaris, and hollows with lawns of S. fallax, Carex rostrata and A. polifolia
114
characterized the sampling area. The terms ―fen‖ and ―bog‖ are used for simplicity and to
115
denote the existence of a trophic and wetness gradient inferred from the vegetation.
116
Environmental manipulations and data collection
117
In each of the two sampling areas, six plots were selected in representative surfaces.
118
Among the 12 sampling plots, the maximal distance between the two most distant plots was
6
ca. 30 m. In both sampling areas, 3 plots (replicates) were randomly assigned as controls and
120
3 plots were assigned as climate warming treatment (begin April, 2008). An increase of air
121
and soil temperatures was passively achieved by placing hexagonal ITEX open-top chambers
122
(hereafter ―OTC‖) over the vegetation (Marion et al., 1997). Since warming in OTC chambers
123
also affects the top-soil humidity, we hereafter name this treatment ―climate effect‖.
124
Hexagonal OTCs were 50 cm high, had a diameter of 1.8 m at the top and 2.5 m at the
125
bottom, and were made of transparent polycarbonate. To reduce edge effects such as reduced
126
precipitation in the chamber we used the OTC design described by Aerts et al. (2004) and
127
Dorrepaal et al. (2004). In each plot, air temperature (10 cm above the Sphagnum surface) and
128
soil temperature (7 cm below the Sphagnum surface) were recorded continuously every 30
129
minutes using thermocouple probes and a datalogger (CR-1000 Campbell). Moreover, in each
130
plot, pH, conductivity, water content of Sphagnum and the depth to the water table (DWT)
131
were measured at each sampling campaign.
132
Every month from 25th May 2009 to 25th November 2009, samples of S. fallax were
133
collected in each plot for the study of phenolic compounds, fungi-producing phenoloxidases
134
and phenoloxidase activities around 10 permanent markers inserted in moss carpets. The goals
135
of this sampling design were (1) to allow for multiple sampling at the site over time, and (2)
136
to obtain a composite sample from each plot and avoid any bias due to spatial heterogeneity.
137
S. fallax shoots were cut into two pieces (sampling depth): 0-3 cm (living ―top segments‖) and
138
3-10 cm (early declining ―bottom segments‖) from the capitulum.
139
Phenolic compounds quantification
140
Primarily bound (hereafter ―bound phenolics‖) and water-soluble phenolic (hereafter ―free
141
phenolics‖) compounds were extracted from lyophilized mosses as described in Jassey et al.
142
(2011a). Briefly, bound phenolic compounds were extracted using ethanol / distilled water
7
solution (80/20 v/v) and free phenolics using distilled water. Free and bound total phenolic
144
contents were quantified with the Folin-Ciocalteau reagent and were expressed in mg
145
equivalent gallic acid (A760) per gram of Sphagnum dry mass (mg g-1 DM). 146
Quantification of culturable fungi-producing phenoloxidases
147
Culturable fungi-producing phenoloxidases were counted as described by Criquet et al.
148
(2000). Two grams fresh weight of Sphagnum was powdered (< 0.5 mm; SEB®Optimo
149
compact mixer) and suspended in 250 mL of a 0.85% NaCl solution with 0.05% Tween 80.
150
This mixture was agitated for 2h on a reciprocal shaker (120 rpm). The extract was diluted
151
(10-1 to 10-3) in NaCl (0.85%) solution and 0.1 mL of each dilution was used to inoculate a
152
medium containing 5 g of malt (Sigma), 15 g of agar (Sigma), 50 mg of chloramphenicol
153
(Sigma) and 0.5 mL of guaiacol (Sigma) per liter. The fungi-producing phenoloxidases were
154
revealed by the red color of the environment related to the oxidation of guaiacol. Results are
155
expressed in colony forming units per gram of Sphagnum dry mass (CFU g-1 DM).
156
Phenoloxidase activities quantification
157
Phenoloxidase activities were quantified following the method described by Criquet et
158
al. (1999). Phenoloxidases were extracted by adding in a Pyrex bottle 3 g of fresh weight of
159
powdered Sphagnum with 50 mL of a 0.1 M CaCl2 solution with 0.05% Tween 80 and 20 g of
160
polyvinylpolypyrrolidone. The samples were shacked at room temperature for 1h on a
161
reciprocal shaker (120 rpm). The suspension of each extract was filtered through a double
162
layer of gauze to remove floating debris and centrifuged at 10 000 g for 10 min at 4°C. Then
163
the supernatant was filtrated through 1.2 µm Whatman GF / D filters and concentrated for 24h
164
in a cellulose-dialysis tube (Medicell Internationel Ltd.) with a 10 kDa molecular mass
cut-165
off, covered with polyethylene glycol (PEG, Sigma-Aldrich), until a final volume of 1/10 of
166
the initial volume. Enzymatic activities were measured using a 96 well microtiter plate with
8 L-DOPA (10 mM). For each sample, 8 pseudo-replicate wells were included. Assay wells
168
received 150 l of extract. Phenoloxidase activities were measured by adding 100 µL of L
-169
DOPA. For each sample, 8 pseudo-replicate wells containing 150 l of boiled extract (2h at
170
90°C) were performed as control. Then samples were incubated at 23°C and L-DOPA
171
oxidation rates were monitored spectrophotometrically at 460 nm for 24h using a microstation
172
plate reader (Bioadvance).
173
Enzymatic activities were calculated by subtracting the mean absorbance of control
174
wells from the mean absorbance of extract wells and by using Beers Law. The molar
175
absorbancy coefficient for the L-DOPA product 3-dihydroindole-5,6-quinone-2-carboxylate
176
(dicq) (3.7 x 104 M-1.cm-1; Mason, 1948) was used and activities were expressed in enzymatic
177
units (U) defined as one nmol of substrate oxidized per h-1 per g of dry m ass.
178
Numerical analysis
179
To compare the general effects of the OTCs on environmental parameters during the 7 months
180
of our study, daily average temperature, as well as minimum and maximum daily
181
temperatures, pH and conductivity were calculated for spring (May-June), summer
(late-June-182
September) and autumn (late-September-November). Then repeated measures ANOVA were
183
computed among sampling areas to focus on the effect of OTCs on these factors with time as
184
repeated measure (time = 3: spring, summer and autumn). The depth and climatic effect on
185
phenolic compounds (free and bound), culturable fungi-producing phenoloxidases and
186
phenoloxidase activities were also analysed using repeated measures ANOVA with time as
187
repeated measure (time = 7: May-November). Each dataset was thereafter split by month to
188
get one response matrix per month for each biological factor using one-way ANOVA. In
189
parallel, correlations between free phenolics, fungi-producing phenoloxidases and
190
phenoloxidase activity in controls and OTCs were determined along the fen-bog gradient
9
using general linear models (GLM) and one-way ANOVAs. The residuals from ANOVAs
192
were tested for normality. Moreover, the coefficient of determination of each variable in the
193
models (adjusted R²) was determined with an analysis of variance.
194
Redundancy analyses (RDA) were applied to Sphagnum related biochemical variables
195
(polyphenols, phenoloxidases, culturable fungi-producing phenoloxidases) for each
196
Sphagnum segment among the fen and the bog areas using climatic treatment (a binary
197
variable with two levels: Control and OTC), Sphagnum moisture content and time (months
198
coded as classes) as explanatory variables. The interactions between climatic treatment and
199
Sphagnum moisture content were also included in the model. The significance of the model
200
and of each explanatory variable included in the model was tested using 1,000 permutations
201
(Gillet et al., 2010). Partial RDAs were also computed after removing the time effect
202
(months) from the ordination following the same method. Additionally, variation partitioning
203
using RDA and adjusted R² was applied to compare the respective effect of each explanatory
204
variable alone (Peres-Neto et al. 2006).
205
Multiple factor analysis (MFA) was used to symmetrically link seven groups of
206
descriptors split in seven sub-matrices: the two Sphagnum related biochemical matrices
207
(phenolic compounds and phenoloxidase data sets), the two abiotic data sets describing
208
physical (depth to water table, air and soil temperature, rainfall and Sphagnum moisture
209
content) and chemical (conductivity and pH) environmental conditions, the climatic data set
210
describing climate treatment (a binary variable with two levels: OTC coded with 1 and control
211
with 0), and the two data sets describing the seasons (spring, summer or autumn coded as
212
classes), and the sampling areas (fen or bog coded as classes). MFA was chosen because it
213
allows the simultaneous coupling of several groups or subsets of variables defined on the
214
same objects and to assess the general structure of the data (Escofier & Pagès, 1994). Briefly,
215
MFA is basically a PCA applied to the whole set of variables in which each subset is
10
weighted, which balances inertia between the different groups and thus balances their
217
influences. RV-coefficients (Pearson correlation coefficient, ranging from 0 to 1) were used to
218
measure the similarities between two data matrices and were tested by permutations (Robert
219
& Escoufier, 1976; Josse et al., 2008). Euclidean distances of global PCA were used in MFA
220
to perform cluster analysis according to the Ward method, and the resulting dendrogram was
221
projected in the MFA ordination space. This allows discovering the main discontinuities
222
among groups and/or sites described by all biotic and abiotic subsets of variables (Carlson et
223
al., 2010; Borcard et al., 2011).
224
All multivariate analyses were performed with the software R 2.10.1 (R Development
225
Core Team 2010) using the vegan (Oksanen et al., 2010) and FactoMineR (Husson et al.,
226 2009) packages. 227 228 Results 229
Seasonal variation of climate variables
230
In spring and summer (May to September), the OTCs significantly increased the daily
231
maximum air temperature (an average of 3°C; ANOVA P < 0.01) and the average air
232
temperature (an average of 1°C; ANOVA P < 0.01). Climate treatment also significantly
233
affected the daily soil temperature in spring in the bog area (an average increase of 0.6°C;
234
ANOVA P < 0.05) and in summer in the fen area (an average increase of 0.8°C; ANOVA P <
235
0.05). No significant differences emerged for the minimum and maximum soil temperatures.
236
In autumn, no significant effect of OTCs was recorded along the gradient for air and soil
237
temperature. An indirect effect of climate treatment was also observed in Sphagnum mosses,
238
since a significant decrease of Sphagnum water content in OTCs was recorded in summer
11
(August and September) in both Sphagnum segments in the bog area, and in top segments in
240
the fen area (ANOVA P < 0.05. Fig. 1).
241
Rainfall significantly varied following the seasons with a decrease from June (156
242
mm) to August, September and October (a monthly average of 72 mm) and an increase in
243
November (231 mm). These variations were also reflected in the depth to water table.
244
Following the seasons and climate treatments, average monthly pH did not significantly vary
245
in both sampling areas (Table 1). Conversely, the conductivity increased from spring to
246
autumn in both sampling areas, with significant differences between controls and OTCs in
247
summer (bog area, P = 0.05) and in autumn (fen area, P = 0.01).
248
Climate effect on phenolic compounds and seasonal variations
249
Regardless of seasonal variations, climate effect and fen-bog gradient, bound and free
250
phenolic contents were significantly higher (ANOVA P < 0.001) in top segments as compared
251
to bottom segments (Fig. 2), except bound phenolics in the bog area (P = 0.16). The two
252
phenolics variables were also positively correlated, with respectively r = 0.38 and 0.37 in the
253
bog area (ANOVA, P < 0.01) and r = 0.70 and 0.41 in the fen area (ANOVA, P < 0.001). The
254
climate effect on bound phenolics resulted in a decrease of concentration of an average of 0.4
255
mg g-1 DM in the two sampling areas, particularly in spring and summer in top segments (P =
256
0.04 and 0.02, respectively). The climate effect on free phenolics was essentially recorded in
257
the fen area for both Sphagnum segments, with constantly lower concentrations in OTCs than
258
in controls over the seasons (ANOVA, P = 0.001) (Fig. 2), whereas the climate effect in the
259
bog area was more rare.
260
In controls, seasonal variations of bound phenolics were recorded in top segments
261
along the fen-bog gradient (P = 0.04 and 0.05, respectively) (Fig. 2a, b, c, d), especially from
262
May to August with a significant decrease of an average of 1.5 mg.g-1 DM. In bottom
12
segments of controls, no significant seasonal variations of bound phenolics were recorded
264
along the fen-bog gradient (P = 0.86 and 0.66, respectively), with an average of respectively
265
1.5 mg g-1 DM in the bog area and 1.0 mg g-1 DM in the fen area. As for bound phenolics,
266
seasonal variations of free phenolics in controls were recorded in top segments with a
267
significant decrease in summer (from 1.4 to 0.8 mg g-1 DM in the two sampling areas; P <
268
0.01 and 0.03, respectively). In bottom segments, no seasonal variations of free phenolics
269
were recorded, with an average of 0.8 mg g-1 DM along the fen-bog gradient (Fig. 2e, f, g, h).
270
In addition, a significant correlation was found between the decrease of phenolics (free and
271
bound) and the decrease of Sphagnum moisture content in summer (ANOVA, P < 0.01) in
272
both segments in the bog area, and in top segments in the fen area.
273
In OTCs, the same seasonal variations as in controls were recorded in Sphagnum
274
segments and for both phenolics along the fen-bog gradient (P < 0.05 for all) (Fig. 2). As for
275
controls, the same significant correlations were recorded between the decrease of phenolics
276
(free and bound) and the decrease of Sphagnum moisture content in summer (ANOVA, P <
277
0.05).
278
Climate effect on culturable fungi-producing phenoloxidases and enzymatic activity, and
279
seasonal variations
280
Significant differences between top and bottom segments of Sphagnum were recorded with
281
overall higher densities of fungi-producing phenoloxidases and higher phenoloxidase
282
activities in bottom segments as compared to top segments in both sampling areas (ANOVA
283
P < 0.05).
284
For densities of culturable fungi-producing phenoloxidases, the climate effect was
285
only significant in the fen area in top segments (ANOVA P = 0.03), with a significant lower
286
value in June in OTCs compared to control (Fig. 3a, b). Seasonal variations were recorded for
13
both Sphagnum segments in the fen and bog area, with a peak in June in controls (P < 0.05)
288
(Fig. 3 a, b, c, d), while in OTCs this peak was only recorded in the bog area (Fig. 3c, d).
289
Climate effects on phenoloxidase activity demonstrated equivocal results in the fen area,
290
while phenoloxidase activity tended to be higher in OTCs in the bog area (Fig. 3e, g).
291
Significant positive correlations were also found between densities of fungi-producing
292
phenoloxidases and extracellular phenoloxidase activities, in both sampling areas and both
293
climate treatments (on average r = 0.40; ANOVA, P < 0.05). In parallel, significant negative
294
correlations between free phenolic compounds and phenoloxidase activities were found for
295
controls in the fen and bog areas when top and bottom Sphagnum segments were pooled (Fig.
296
4a, b). The same tendency was recorded in OTCs, except in the bog area (Fig. 4b).
297
Additionally, the combination of fungi and free phenols in a general linear model explained
298
respectively 27.4% and 10.6% of the variability of phenoloxidase activity in controls, and
299
29.6% and 0.6% in OTCs in the bog area (adjusted R2; P < 0.001). For the fen area another
300
patterns occurred since fungi and free phenolics explained respectively 13.7% and 9.8% of the
301
variability of phenoloxidase activity in controls, and 11.3% and 25.8% in OTCs (adjusted R2;
302
P < 0.001).
303
The phenol-phenoloxidase complex and its relation to abiotic variables
304
The contribution of the explanatory variables in the RDA (Table. 2) showed that time
305
(months) has a major influence on the moss biochemical patterns. In bottom segments
306
sampling time explained between 41% and 66% of the variation. In top moss segments,
307
biplots of partial RDAs showed that Sphagnum related biochemical variables were influenced
308
by climate treatment, as shown by the separation of control and OTC plots along the first
309
RDA axis (Fig. 5a, c). Together, OTCs and Sphagnum moisture content explained 20.6%
310
(fen) and 27.1% (bog) of the variation of biochemical factors (P < 0.05) in top segments.
14
Variation partitioning and adjusted R² showed that OTCs alone explained a higher variation in
312
the fen area than in the bog area, whereas Sphagnum moisture content has higher influence in
313
the bog than in the fen (Table 2). On the other hand, the biochemical descriptors showed a
314
strong opposition between phenolics and warming treatment (OTC) in all biplots, while fungi
315
appears linked to Sphagnum moisture content, particularly in top segments.
316
If we consider all samples together along the fen-bog gradient in the multiple factor
317
analysis (Fig. 6), a clear pattern appeared, with a split into the three seasons (spring, summer
318
and autumn) and within each partition a subdivision into fen and bog areas, each of these
319
subdivisions being further divided into OTC and control plots. The RV-coefficients (Table 3)
320
indicate strongest links between Sphagnum related biochemical variables, sampling area,
321
climate warming and seasons, and between sampling area and physicochemical environment.
322
323
Discussion 324
Polyphenol/phenoloxidase interplay in Sphagnum mosses and along the fen-bog gradient
325
Sphagnum related biochemical factors quantified in this work yielded different results
326
according to Sphagnum segments. Total phenolic content (free and bound) was higher in
327
living top segments as compared to decaying bottom segments in both sampling areas. Such
328
differences have been also observed in S. fallax under controlled conditions (Jassey et al.,
329
2011b). This phenomenon is explained by a higher phenolic metabolism in capitulum than in
330
lower part of the shoot, since Sphagnum capitula (top segments) constitute the living part of
331
the moss where most of the metabolic processes occur, including the growth (Clymo &
332
Hayward, 1982). The reduction of phenolics towards the lower part of the shoot was also
333
accompanied by an increase of culturable fungi-producing phenoloxidases and of
334
phenoloxidase activity, suggesting a higher degradation of recalcitrant phenolics in early
15
declining Sphagnum segments (Baldrian, 2006; Toberman et al., 2010; Sinsabaugh & Follstad
336
Shah, 2011). These results also pointed to the fact that at low concentrations free phenols may
337
induce phenoloxidase activity, and inhibit the oxidation activity at high concentration
338
(Sinsabaugh, 2010). Given that no clear correlation was found between fungi and free
339
phenols, such vertical gradient also highlighted a possible direct inhibitory effect of free
340
phenols on phenoloxidase activity (Wetzel, 1992; Freeman et al., 2001a; Fenner et al., 2005).
341
Our results likewise demonstrated a strong relationship between fungi and
342
phenoloxidase activities. Phenoloxidase activity is essentially attributable to lignolytic fungi
343
such as basidiomycetes (Criquet et al., 2000; Thormann et al., 2002; Baldrian, 2006). Fungal
344
activity is known to be directly influenced by the supply of organic matter (Berg et al., 1998;
345
Criquet et al., 2000). A study in the same experimental site demonstrated over the fen-bog
346
gradient an increase of organic matter content in the upper 10 cm soil layer, which induced
347
higher fungal activity (Delarue et al., 2011). Thus, all of these findings emphasize that
348
phenoloxidase activity was mainly controlled by fungi and secondarily by phenols.
349
Beside the differences between Sphagnum segments, different patterns of polyphenol
350
content and phenoloxidase activities were recorded along the fen-bog gradient over the
351
seasons. In particular, phenoloxidase activities were more intense in the bog area than in the
352
fen area. Again, this result appeared linked to fungi. The abundance of vascular plants is
353
higher in the bog area and supplies more easily decomposable organic matter, favouring
354
fungal activity (Delarue et al., 2011). A number of studies have demonstrated that fen and bog
355
litters were characterized by distinct patterns of microfungal community, especially in the
356
surface horizons (Thormann et al., 2001, 2002, 2004; Thormann, 2006; Artz et al., 2007).
357
Thus, vegetation patchiness along the fen-bog gradient may directly affect fungal community
358
composition, and indirectly phenoloxidase activity. In particular, the quality and quantity of
359
plant-derived labile carbon resulting from vegetation succession may directly influence fungal
16
diversity, e.g. polymer- and recalcitrant polymer degraders (Thormann, 2006). On the other
361
hand, the influence of free phenols on phenoloxidases was higher in the fen area than in the
362
bog and this could be explained by qualitative differences of phenolics in Sphagnum along the
363
gradient (Opelt et al., 2007). When comparing phenolic content in Sphagnum from different
364
ecological setting, Folin assay only gives a global tendency of phenolic variation, and not the
365
quality of free phenols that may influence phenoloxidase activity. Such results clearly call for
366
a detailed analysis of phenolic variation (e.g. phenolic acids or flavonoids).
367
Climate effect on polyphenols, phenoloxidases and their interactions along the fen-bog
368
gradient
369
As described in previous studies (Dorrepaal et al., 2004; Aerts, 2006), higher air temperatures
370
induced higher evapotranspiration, which resulted in lower Sphagnum moisture content
371
during summertime. Obviously, higher evapotranspiration also could have sometimes induced
372
lower soil temperature by heat loss towards atmosphere and reduction of soil thermal
373
conductivity, thus explaining the so-called marginal effect of OTCs on soil temperature
374
(Dabros et al., 2010). Despite contrasted effects of OTCs on air and soil temperature, a
375
climate effect has been recorded on biochemical variables measured along Sphagnum
376
segments.
377
Seasonal effects were predominant for the biochemical variation in Sphagnum carpet.
378
However, multivariate analyses revealed a climate warming effect beyond the seasonal
379
variations of Sphagnum biochemical related factors. As observed elsewhere (Aerts, 2006;
380
Bragazza, 2008; Dabros and Fyles, 2010; Dabros et al., 2010), the increase of air temperature
381
associated with the reduction in rainfall led to heat waves, and the impact of these events was
382
exacerbated in OTCs increasing drought in top-soil. Enhanced top-soil aeration as a result of
383
water table drawdown and air temperature increase was recognized to influence
17
phenoloxidase activity and polyphenols (Freeman et al., 1993, 2001a, b; Toberman et al.,
385
2008; Ellis et al., 2009). As supported by current findings in peatlands (Pind et al., 1994;
386
Williams et al., 2000; Freeman et al., 2001a; Toberman et al., 2008, 2010; Sinsabaugh, 2010),
387
peat soil environmental factors (i.e. acidic pH, water table depth, and oxygen) mainly inhibit
388
phenoloxidase activity, explaining our weak variations of phenoloxidases with climate
389
warming.
390
In parallel, climate warming had greatest impact on the phenolic metabolism with a
391
decrease of phenolics related to the decrease of Sphagnum moisture in OTCs and the increase
392
of air temperatures. The level of total phenolic compounds tends to be lower in several boreal
393
species under elevated temperatures (Veteli et al., 2007). Such decrease may be explained by
394
a diminution of carbon partitioning to phenolics (Herms & Mattson, 1992; Mattson et al.,
395
2005). Elevated temperatures are recognized to induce better growth of Sphagnum species
396
(Breeuwer et al., 2008). It might well be that a trade-off between growth and differentiation
397
(i.e. the production of carbon-based secondary metabolites such as phenols) occurred, with a
398
potential diminution of carbon skeletons allocation to phenolics (Mattson et al., 2005; Veteli
399
et al., 2007). Such results imply that any repeated significant decrease of phenolics through
400
more intense and frequent heat waves − as predicted by climate scenarios (Meehl & Tabaldi,
401
2004; Schär et al., 2004; IPCC, 2007) − will probably lead to the opening of the enzymatic
402
latch, as described by Freeman et al. (2001b).
403
Furthermore, our climate experiment demonstrated that climate warming has not had
404
the same impact along the fen-bog gradient since a stronger decrease of polyphenols was
405
recorded in the fen area. This decrease induced a switch between fungi and free phenols,
406
leading to a reduction of the potential inhibitory effect of free phenols on phenoloxidases.
407
However, the decrease in the density of culturable fungi-producing phenoloxidase during
408
dryer periods could not compensate for the decrease of phenolics and lowering of their
18
inhibitory effect on phenoloxidase activity. Alternatively, or additionally, phenolics may also
410
have inhibitory effects on other microbial activities with implication for the carbon cycle,
411
such as hydrolase activity (Fenner et al., 2005, 2007). Thus, the reduction of the inhibitory
412
effect of free phenols could affect carbon cycling in the fen area through another
413
microbial/polyphenols interplay (e.g. Jassey et al., 2011a). In the bog area phenoloxidase
414
activity remained the key factor influenced by climate treatment with a slight increase of
415
activity in top segments, leading to potentially higher degradation of recalcitrant materials in
416
surface horizons. In contrast to the fen area, it appeared that fungi mainly influenced
417
phenoloxidases in OTCs, as shown by GLMs.
418
Although a slight increase of temperature induced by OTCs is not strong enough to
419
significantly affect the decomposition rate of Sphagnum litter on short-time scale (Dabros et
420
al., 2010), our results demonstrated that already within a 7-month period key elements of the
421
carbon cycle can be altered in surface horizons. Furthermore, our climate experiment
422
highlights different responses of Sphagnum related biochemical variables along the fen-bog
423
gradient. The main consequence is that not all the peatland habitats would respond similarly
424
to climate forcing. Ultimately, our results suggest a destabilization of peatland ecosystems
425
and reinforce the point that phenoloxidase/polyphenol interplay is especially critical to
426
understanding the response of peatlands to climate change.
427
428
Acknowledgments 429
This research is a contribution of the ANR PEATWARM project (Effect of moderate
430
warming on the functioning of Sphagnum peatlands and their function as a carbon sink).
431
PEATWARM is supported by the French National Agency for Research under the
432
‗‗Vulnerability: Environment—Climate‘‘ Program (ANR-07-VUL-010). Further funding to
433
V. Jassey by the Franche-Comté Region is kindly acknowledged. The authors would like to
19
thank F. Gillet (Université de Franche-Comté, France) for his statistical assistance. They also
435
thank R. Payne (University of Manchester, England) for his English edits, and the three
436
reviewers for their valuable review of this work.
437
438
References 439
Aerts R (2006) The freezer defrosting: global warming and litter decomposition rates in cold
440
biomes. Journal of Ecology 94(4): 713-724.
441
Aerts R, Cornelissen JHC, Dorrepaal E, van Logtestijn RSP, Callaghan TV (2004) Effects of
442
experimentally imposed climate scenarios on flowering phenology and flower
443
production of subarctic bog species. Global Change Biology 10(9): 1599-1609.
444
Artz REE, Anderson IC, Chapman SJ, Hagn A, Schloter M, Potts JM, Campbell CD (2007)
445
Changes in fungal community composition in response to vegetational succession
446
during the natural regeneration of cutover peatlands. Microbial Ecology 54(3):
447
508:522.
448
Baldrian P (2006) Fungal laccases - occurrence and properties. Fems Microbiology Reviews
449
30(2): 215-242. 450
Berg MP, Kniese JP, Verhoef HA (1998) Dynamics and stratification of bacteria and fungi in
451
the organic layers of a Scots pine forest soil. Biology and Fertility of Soils 26(4):
313-452
322.
453
Borcard D, Gillet F, Legendre P (2011) Numerical Ecology with R. Use R! Series, Springer,
454
New York NY, USA.X. ISBN: 978-1-4419-7975-9.
455
Bragazza L (2008) A climatic threshold triggers the die-off of peat mosses during an extreme
456
heat wave. Global Change Biology 14(11): 2688-2695.
457
Breeuwer A, Heijmans M, Robroek BJM, Berendse F (2008) The effect of temperature on
458
growth and competition between Sphagnum species. Oecologia 156(1): 155-167.
459
Bridgham SD, Updegraff K, Pastor J (1998) Carbon, nitrogen, and phosphorus mineralization
460
in northern wetlands. Ecology 79(7): 2571-2571.
461
Bridgham SD, Updegraff K, Pastor J (2001) A comparison of nutrient availability indices
462
along an ombrotrophic-minerotrophic gradient in Minnesota wetlands. Soil Science
463
Society of America Journal 65(1): 259-269.
464
Carlson ML, Flagstad LA, Gillet F, Mitchell EAD (2010) Community development along a
465
proglacial chronosequence: are above-ground and below-ground community structure
466
controlled more by biotic than abiotic factors? Journal of Ecology 98(5): 1084-1095.
467
Clymo RS, Hayward PM (1982) The ecology of Sphagnum. In: Bryophyte Ecology. AEJ
468
Smith. Chapman & Hall, New York. pp. 229-289.
469
Criquet S, Farnet AM, Tagger S, Le Petit J (2000) Annual variations of phenoloxidase
470
activities in an evergreen oak litter: influence of certain biotic and abiotic factors. Soil
471
Biology & Biochemistry 32(11-12): 1505-1513.
20
Criquet S, Tagger S, Vogt G, Iacazio G, Le Petit J (1999) Laccase activity of forest litter. Soil
473
Biology & Biochemistry 31(9): 1239-1244.
474
Dabros A, Fyles JW (2010) Effects of open-top chambers and substrate type on
475
biogeochemical processes at disturbed boreal forest sites in northwestern Quebec.
476
Plant and Soil 327(1-2): 465-479.
477
Dabros A, Fyles JW, Strachan IB (2010) Effects of open-top chambers on physical properties
478
of air and soil at post-disturbance sites in northwestern Quebec. Plant and Soil
333(1-479
2): 203-218.
480
Delarue F, Laggoun-Défarge F, Disnar JR, Lottier N, Gogo S (2011) Organic matter sources
481
and decay assessment in a Sphagnum-dominated peatland (Le Forbonnet, Jura
482
Mountains, France): impact of moisture conditions. Biogeochemistry (in press).
483
Dorrepaal E, Aerts R, Cornelissen JHC, Callaghan TV, van Logtestijn RSP (2004) Summer
484
warming and increased winter snow cover affect Sphagnum fuscum growth, structure
485
and production in a sub-arctic bog. Global Change Biology 10(1): 93-104.
486
Ellis T, Hill PW, Fenner N, Williams GG, Godbold D, Freeman C (2009) The interactive
487
effects of elevated carbon dioxide and water table draw-down on carbon cycling in a
488
Welsh ombrotrophic bog. Ecological Engineering 35(6): 978-986.
489
Escofier B, Pages J (1994) Multiple factor-analysis (afmult package). Computational
490
Statistics & Data Analysis 18(1): 121-140.
491
Fenner N, Freeman C, Reynolds B (2005) Hydrological effects on the diversity of phenolic
492
degrading bacteria in a peatland: implications for carbon cycling. Soil Biology &
493
Biochemistry 37(7): 1277-1287.
494
Fenner N, Ostle NJ, McNamara N, Sparks T, Harmens H, Reynolds B, Freeman C (2007)
495
Elevated CO2 effects on peatland plant community carbon dynamics and DOC
496
production. Ecosystems 10(4): 635-647.
497
Freeman C, Evans CD, Monteith DT, Reynolds B, Fenner N (2001a) Export of organic
498
carbon from peat soils. Nature 412(6849): 785-785.
499
Freeman C, Lock MA, Reynolds B (1993) Fluxes of CO2, CH4 and N2O from a welsh
500
peatland following simulation of water-table draw-down - potential feedback to
501
climatic-change. Biogeochemistry 19(1): 51-60.
502
Freeman C, Ostle N, Kang H (2001b) An enzymic 'latch' on a global carbon store - A shortage
503
of oxygen locks up carbon in peatlands by restraining a single enzyme. Nature
504
409(6817): 149-149. 505
Freeman C, Ostle NJ, Fenner N, Kang H (2004) A regulatory role for phenol oxidase during
506
decomposition in peatlands. Soil Biology & Biochemistry 36(10): 1663-1667.
507
Gillet F, Peter M, Ayer F, Butler R, Egli S (2010) Long-term dynamics of aboveground
508
fungal communities in a subalpine Norway spruce forest under elevated nitrogen
509
input. Oecologia 164(2): 499-510.
510
Gorham E (1991) Northern peatlands: role in the carbon cycle and probable responses to
511
climatic warming. Ecological Applications 1(2): 181-195.
512
Herms DA, Mattson WJ (1992) The dilemma of plants - to grow or defend. Quarterly Review
513
of Biology 67(3): 283-335.
514
Husson F, Josse J, Lê S, Mazet J 2009. FactoMineR: Factor Analysis and Data Mining with
21
R.In: R package, version 1.12 http://CRAN.R-project.org/package=FactoMineR.
516
IPCC (2007) Solomon S, Qin D, Manning M, Chen Z, Marquis M, Averyt KB, Tignor M,
517
Miller HL, eds. Climate change 2007: the Physical Science Basis. Contribution of
518
Working Group I to the Fourth Assessment Report of the Intergovernmental Panel on
519
Climate Change. Cambridge University Press, Cambridge, UK and New York, NY,
520
USA, p 996
521
Jassey VEJ, Chiapusio G, Mitchell EAD, Binet P, Toussaint ML, Gilbert D (2011a)
Fine-522
scale horizontal and vertical micro-distribution patterns of testate amoebae along a
523
narrow fen/bog gradient. Microbial Ecology 61(2) 374-385.
524
Jassey VEJ, Gilbert D, Binet P, Toussaint M-L, Chiapusio G (2011b) Effect of a temperature
525
gradient on Sphagnum fallax and its associated microbial communities: a study under
526
controlled conditions. Canadian Journal of Microbiology 57(3) 226-235.
527
Josse J, Pages J, Husson F (2008) Testing the significance of the RV coefficient.
528
Computational Statistics & Data Analysis 53(1): 82-91.
529
Laiho R (2006) Decomposition in peatlands: Reconciling seemingly contrasting results on the
530
impacts of lowered water levels. Soil Biology & Biochemistry 38(8): 2011-2024.
531
Marion GM, Henry GHR, Freckman DW, Johnstone J, Jones G, Jones MH, Levesque E,
532
Molau U, Molgaard P, Parsons AN, Svoboda J, Virginia RA (1997) Open-top designs
533
for manipulating field temperature in high-latitude ecosystems. Global Change
534
Biology 3: 20-32.
535
Mason HS (1948) The chemistry of melanin III. Mechanism of the oxidation of
536
dihydroxypehnylalanine by tyrosinase. Journal Of Biological Chemistry 172 83-99.
537
Mattson WJ, Julkunen-Tiitto R, Herms DA (2005) CO2 enrichment and carbon partitioning to
538
phenolics: do plant responses accord better with the protein competition or the growth
539
differentiation balance models? Oikos 111(2): 337-347.
540
Meehl GA, Tebaldi C (2004) More intense, more frequent, and longer lasting heat waves in
541
the 21st century. Science 305(5686): 994-997.
542
Mellegard H, Stalheim T, Hormazabal V, Granum PE, Hardy SP (2009) Antibacterial activity
543
of sphagnum acid and other phenolic compounds found in Sphagnum papillosum
544
against food-borne bacteria. Letters in Applied Microbiology 49(1): 85-90.
545
Moore PD (2002) The future of cool temperate bogs. Environmental Conservation 29(1):
3-546
20.
547
Oksanen J, Blanchet G, Kindt R, Legendre P, O'Hara RG, Simpson GL, Solymos P, Stevens
548
MHH, Wagner H (2010) vegan: Community Ecology Package. R package version
549
1.17-1. http://CRAN.R-project.org/package=vegan
550
Opelt K, Chobot V, Hadacek F, Schonmann S, Eberl L, Berg G (2007) Investigations of the
551
structure and function of bacterial communities associated with Sphagnum mosses.
552
Environmental Microbiology 9(11): 2795-2809.
553
Peres-Neto PR, Legendre P, Dray S, Borcard D (2006) Variation partitioning of species data
554
matrices: Estimation and comparison of fractions. Ecology 87: 2614-2625
555
Pind A, Freeman C, Lock MA (1994) Enzymatic degradation of phenolic materials in
556
peatlands-measurment of phenol oxidase activity. Plant and Soil 159(2): 227-231.
557
R Development Core Team (2010) R: A language and environment for statistical computing.
22
R Foundation for Statistical Computing, Vienna, Austria. ISBN 3-900051-07-0, URL
559
http://www.R-project.org.
560
Robert P, Escoufier Y (1976) Unifying tool for linear multivariate statistical-methods -
rv-561
coefficient. Journal of the Royal Statistical Society Series C-Applied Statistics 25(3):
562
257-265.
563
Rydin H, Jeglum JK (2006) The Biology of peatlands. In: Oxford University Press. p 354.
564
Schar C, Vidale PL, Luthi D, Frei C, Haberli C, Liniger MA, Appenzeller C (2004) The role
565
of increasing temperature variability in European summer heatwaves. Nature
566
427(6972): 332-336. 567
Sinsabaugh RL (2010) Phenol oxidase, peroxidase and organic matter dynamics of soil. Soil
568
Biology & Biochemistry 42(3): 391-404.
569
Sinsabaugh RL, Shah JJF (2011) Ecoenzymatic stoichiometry of recalcitrant organic matter
570
decomposition: the growth rate hypothesis in reverse. Biogeochemistry 102(1-3):
31-571
43.
572
Smith LC, MacDonald GM, Velichko AA, Beilman DW, Borisova OK, Frey KE,
573
Kremenetski KV, Sheng Y (2004) Siberian peatlands a net carbon sink and global
574
methane source since the early Holocene. Science 303(5656): 353-356.
575
Strack M (2008) Peatlands and Climate Change. In: International Peat Society, Vapaudenkatu
576
12, 40100 Jyväskylä, Finland. p 235.
577
Thormann MN (2006) Diversity and function of fungi in peatlands: A carbon cycling
578
perspective. Canadian Journal of Soil Science 86(2): 281-293.
579
Thormann MN, Currah RS, Bayley SE (2001) Microfungi isolated from Sphagnum fuscum
580
from a southern boreal bog in Alberta, Canada. Bryologist 104(4): 548-559.
581
Thormann MN, Currah RS, Bayley SE (2002) The relative ability of fungi from Sphagnum
582
fuscum to decompose selected carbon substrates. Canadian Journal of Microbiology
583
48(3): 204-211. 584
Thormann MN, Currah RS, Bayley SE (2004) Patterns of distribution of microfungi in
585
decomposing bog and fen plants. Canadian Journal of Botany 82(5): 710-720.
586
Toberman H, Freeman C, Evans C, Fenner N, Artz RRE (2008) Summer drought decreases
587
soil fungal diversity and associated phenol oxidase activity in upland Calluna
588
heathland soil. Fems Microbiology Ecology 66(2): 426-436.
589
Toberman H, Laiho R, Evans CD, Artz RRE, Fenner N, Strakova P, Freeman C (2010)
Long-590
term drainage for forestry inhibits extracellular phenol oxidase activity in Finnish
591
boreal mire peat. European Journal of Soil Science 61(6): 950-957.
592
van Breemen N (1995) How Sphagnum Bogs Down Other Plants. Tree 10: 270-275.
593
Verhoeven JTA, Liefveld WM (1997) The ecological significance of organochemical
594
compounds in Sphagnum. Acta Botanica Neerlandica 46(2): 117-130.
595
Verhoeven JTA, Toth E (1995) Decomposition of Carex and Sphagnum litter in fens - effect
596
of litter quality and inhibition by living tissue-homogenates. Soil Biology &
597
Biochemistry 27(3): 271-275.
598
Veteli TO, Mattson WJ, Niemela P, Julkunen-Tiitto R, Kellomaki S, Kuokkanen K, Lavola A
599
(2007) Do elevated temperature and CO2 generally have counteracting effects on
600
phenolic phytochemistry of boreal trees? Journal of Chemical Ecology 33(2):
23
296.
602
Weltzin JF, Bridgham SD, Pastor J, Chen JQ, Harth C (2003) Potential effects of warming
603
and drying on peatland plant community composition. Global Change Biology 9(2):
604
141-151.
605
Weltzin JF, Harth C, Bridgham SD, Pastor J, Vonderharr M (2001) Production and
606
microtopography of bog bryophytes: response to warming and water-table
607
manipulations. Oecologia 128(4): 557-565.
608
Weltzin JF, Pastor J, Harth C, Bridgham SD, Updegraff K, Chapin CT (2000) Response of
609
bog and fen plant communities to warming and water-table manipulations. Ecology
610
81(12): 3464-3478. 611
Wetzel RG (1992) Gradient-dominated ecosystems - sources and regulatory functions of
612
dissolved organic-matter in fresh-water ecosystems. Hydrobiologia 229: 181-198.
613
Wheeler BD, Proctor MCF (2000) Ecological gradients, subdivisions and terminology of
614
north-west European mires. Journal of Ecology 88(2): 187-203.
615
Williams CJ, Shingara EA, Yavitt JB (2000) Phenol oxidase activity in peatlands in New
616
York State: Response to summer drought and peat type. Wetlands 20(2): 416-421
617 618
24
Table 1: Seasonal variations of environmental variables measured in controls and OTCs in the
619
fen and bog sampling areas in Le Forbonnet mire (French Jura). Letters indicate significant
620
seasonal variations (P < 0.05). Asterisks indicate significant variations between controls and
621
OTCs (P < 0.05).
622
Table 2: Summary of RDA on Sphagnum related biochemical variables and environmental
623
explanatory variables from Le Forbonnet mire (French Jura): fraction of variance explained
624
and significance of individual variables taken alone. Sph moisture = Sphagnum moisture
625
content; clim treat = climate treatment.
626
Table 3: RV-coefficients (RV) and corresponding P-values among the six groups of variables
627
used in the Multiple factor analysis (MFA) of the entire data set split into 6 groups of
628
variables describing Sphagnum biochemistry, environmental physical and chemical
629
conditions, climate warming treatment, seasons, depth of moss segment and bog/fen areas .
630
Significant coefficients are in bold.
631 632 633 634 635 636 637 638 639
25
Figures:
640
Figure 1: Seasonal variations of Sphagnum moisture content in the two shoot segments (top
641
and bottom) in controls and OTCs of the fen (a, b) and bog (c, d) areas. Mean ± S.E. (n = 3).
642
Asterisk indicates significant difference between controls and OTCs (ANOVA tests, P <
643
0.05).
644
Figure 2: Seasonal variations of bound (a, b, c, d) and free (e, f, g, h) phenolics in the two
645
shoot segments (top and bottom) in controls and OTCs of the bog and fen areas. Mean ± S.E.
646
(n = 3). Asterisk indicates significant difference between controls and OTCs (ANOVA tests, P
647
< 0.05).
648
Figure 3: Seasonal variations of densities of fungi producing phenoloxidases (a, b, c, d) and
649
phenoloxidase activities (e, f, g, h) in the two shoot segments (top and bottom) in controls and
650
OTCs of the bog and fen areas. Mean ± S.E. (n = 3). Asterisk indicates significant difference
651
between controls and OTCs (ANOVA tests, P < 0.05).
652
Figure 4: Correlations between free phenolics and phenoloxidase activity for Sphagnum
653
segments (top and bottom segments pooled) in controls and OTCs in the fen (a) and bog (b)
654
areas.
655
Figure 5: Biplots of redundancy analyses (RDA) of biochemical data measured on Sphagnum
656
mosses (free and bound phenolics, phenoloxidases and fungi-producing phenoloxidases) in
657
top (a) and bottom (b) Sphagnum segments of the fen area, and in top (c) and bottom (d)
658
segments of the bog area. Climate treatments are coded with open symbol for controls and
659
with filled symbol for OTCs. Months are indicated next to the sample points by their number.
660
Season effect has been removed by giving the variable months as covariable. Environmental
661
variables are represented by vectors (arrows for quantitative or semi-quantitative variables):
662
Sph_moist.: Sphagnum moisture content; Sph_moist:OTC: interactions between Sphagnum
26
moisture and OTCs. Biochemical variables are given with dotted arrows: F_phen: free
664
phenolics, B_phen: bound phenolics; Phen_oxid: phenoloxidase activity; Fungi: culturable
665
fungi-producing phenoloxidase. Axes are significant (P< 0.05), except for bottom segments.
666
Axes 3 are never significant, with less than 1% of variance). Grey ellipses represent S.E. of
667
site scores around the centroid of each treatment level.
668
Figure 6: Multiple factor analysis (MFA) samples biplot of the entire data set split into 7
669
groups of variables describing Sphagnum biochemistry, environmental physical and chemical
670
conditions, climate warming treatment, seasons and fen-bog areas. Biplot of axes 1 and 2
671
(both significant at P = 0.001) is given together with the result of a hierarchical agglomerative
672
clustering (grey lines) obtained by the Ward method on the Euclidean distance matrix
673
between MFA site scores, showing three main groups of sampling plots (circles = spring,
674
squares = summer, triangles = autumn) and two sub-groups (white symbols = controls, black
675
symbols = OTCs). Sampling areas are indicated with letters besides sampling plots (F: fen
676
area; B: bog area).
677
678
679
680
